US2954531A - Transistor oscillator - Google Patents

Description

Sept. 27, 1960 R. P. JOHNSON 2,954,531

TRANSISTOR OSCILLATOR Filed March 4, 1959 |3 START START FINISH FINISH PRIMARY WINDING WOUND NEXT TO CASE START STAET FTNISH FINISH SECONDARY WINDING WOUND OVER PRIMARY 1; o INVENTOR. J g RICHARD F? JOHNSON.

wjim w.

ATTORNEYS.

Patented Sept. 27, 1960 TRANSISTOR OSCILL'ATOR Richard P. Johnson, Hyde Park, Mass, assignor to Avco Manufacturing Corporation, Cincinnati, Ohio, a corporation of Delaware Filed Mar. 4, 19'59, Ser. No. 797,198

2 Claims. (Cl. 331-113) The present invention relates to transistorized power supplies, commonly referred to as transverters, and particularly to a heavy duty high efficiency power supply.

The following United States patents are illustrative of the general background to which the present improve- Certain Workers in the art have observed that the output wave forms of transverters are frequently characterized by voltage spikes which find their way into the low level audio stages of receivers or other equipment powered by the transverter. Further, they destroy transistor junctions.

in accordance with the invention there is provided a transverter featuring a bifilar-wound saturable timing and switching or feedback transformer.

The principal object of this novel transverter, incorporating the aforementioned feature, is to reduce leakage inductance which in turn minimizes the switching transients or voltage spikes referred to above.

Another object of the invention is to provide a separate non-saturable shunt-connected voltage step-up transformer in combination with the saturable timing and switching transformer.

Another object of the invention is to provide a transverter in which the peak-to-peak value of the emitter-toemitter voltage very closely approximates twice the value of the DC). supply voltage.

A further object of the invention is to provide a circuit in which there is no direct current flow in the primary winding of the step-up or output transformer.

Advantages of the invention coextensive with other objects thereof are to provide a mobile power supply having substantial output power, good voltage regulation, and high over-all efficiency.

Yet another object of the invention is to provide a power supply which is inherently self-protecting against overload. The fundamental circuit operation is such that when a short-circuit or heavy overload occurs, the oscillatory action ceases and the input current goes to a low value where it remains until the trouble is corrected.

For a better understanding of the present invention, together with other objects, advantages, and capabilities thereof, reference is made to the appended description of the accompanying drawings, in which there is shown a preferred embodiment of a transverter in accordance with the invention.

In the drawings:

Fig. 1 is a circuit schematic of a transverter in accordance with the invention; and

Fig. 2 is a plan view showing the mode of wiring the primary and secondary of the switching transformer.

Reference is first made to the saturable switching or timing transformer indicated by the reference numeral 10. It comprises a toroidal core 11, primary windings 12 and 13, and secondary windings 14 and 15, bifilarly arranged in the manner shown in Fig. 2. In one Working embodiment of the invention, a nylon-covered toroidal core was employed and it was tightly wrapped with a layer of Scotch electrical tape, each turn being overlapped about half the width of the tape.

The primary is wound first by cutting two lengths of No. 14 Nyclad wire, each 112 inches long. The wires 12 and 13 are held parallel to each other and taped together at their centers. The central mark is placed against the core and the two lengths of wire are wound side by side in both directions away from the center holes, keeping the wires tight, with even spacings and turns distribution so that the entire core is covered. The entire primary is then wound with two layers of Scotch tape.

Following the winding of the primary, two lengths of No. 26 Nyclad wire, each inches long, are cut, and the secondary comprising these wires 14 and 15 is wound directly over the primary in the same direction as the primary. The entire secondary is covered with a single layer of Scotch tape, leaving the leads protruding from the tape. This transformer is incorporated in the Fig. 1 transverter circuit as shown.

Referring now specifically to Fig. 1, two transistors generally indicated by the reference numerals 16 and 17 are connected in the common collector configuration, the collectors being connected at 18 to the negative terminal of the battery or DC. source 19. The saturable switching or feedback transformer 10 has its secondary windings 14 and 15 connected in series between the bases 20 and 21. The primaries 12 and 13 are connected in series between the emitters 22 and 23, and the center tap between the two primary windings is connected at 24 to the positive terminal of the direct current battery or source 19.

One of the difliculties which has been encountered in transverter power supplies is failure to oscillate at low temperatures upon application of direct current input voltage, because of the fact that current gain of the transistors is lowered as the temperature decreases. This is overcome by connecting unequal resistors 25 and 26 across the DC. source and by connecting the junction 27 of these resistors to the center tap 28 between the secondary windings. These resistors bias the two transistors in the forward direction so that there is a small initial current flow from emitter to collector when the input voltage is applied. Any small circuit unbalance causes one transistor initially to conduct more heavily than the other, and since feedback is developed immediately through the transformer 10, oscillations begin immediately.

The emitters are connected to the primary 29 of a non saturated output transformer 30, the secondary 31 of which is connected to a conventional bridge rectifier network comprising diodes 32, 33, 34, and 35. Each diode in the drawing actually represents two Sarkes Tarzian M-SOOs in series in the prototype power supply. The reason for this is that a safety factor is needed on'the peak inverse voltage rating. This network is in turn connected at 37 and 38 to a filter comprising shunt capacitor 39, series inductance 40, and shunt resistor 41, the high voltage being available at 4-2 and 43. A connection is also brought out to output terminal 44 from the center tap 45 of the output transformer, in order to obtain a second output voltage corresponding to one half of the total transformer secondary voltage, via a filter network comprising shunt capacitor 46, series choke 47, and shunt resistor '48.

Optionally, a filter capacitor having a value of 2000 microfarads maybe incorporated in a circuit having the illustrative parameters mentioned below. 'l

I have found the following cireuit'parameters to be adequate in one successful embodiment of the invention: Diodes 32, 33, 34, and 35 Silicon power diodes,

' I 400 volt inverse peak500 milliamperes D.C., Sarkes Tarzian M- 500 or 1Nl084.

Voltage or DC power supply 14 volts.

Maximum total D.C. power output. 140 watts.

Resistor 25 3.3 ohms.

Resistor 26 150 ohms. Transistors Delco type 2N278. Transformer type Arnold Engineering The points of novelty of this circuitry are two. First, the feedback transformer primary and secondary are bifilarly wound. I have found that a novel transverter combination incorporating this feature substantially reduces leakage inductance and overcomes the voltage spike effect With which the art has been seriously troubled.

Second, the shunt-connected output transformer transforms the square-wave voltage existing across the emitters of the switching transistors to a higher-voltage square wave for rectification.

It will be understood that'any suitable rectifier network may be coupled to the output of transformer 30 in lieu of that herein shown.

Static conditions In reference to Fig. 1, consider the voltages present in the circuit before oscillations commence, using point 18 as a reference for voltage measurement. Assume a battery 19 supply voltage of 12 volts.

Point 24, which is the junction of one end; of resistor 25, the finish of primary winding 12, and the start of primary winding 13, is at a potential of +12 volts.

Emitter 22 of transistor 16 and emitter 23 of transistor 17 are connected, through the low resistances of primary windings 12 and 13, respectively, to point 24, and thus are also at a potential. of +12 volts.

Resistors 25 and 26 constitute a bleeder. network which is incorporated in the design to facilitate the start of oscillations. This is accomplished in the following manner.

Due to current flow in the bleeder network and the ratio of resistors 25 and 26, the midpoint.27 of the network is slightly less positive than point 24. The midpoint 27 is connected to, the finish of feedback winding 14 and the start of feedback winding 15 at point 28.

Base 20 of transistor 16 and base 21 of transistor '17 are connected through the low resistances of the feedback windings 14 and 15, respectively, to point 28 and thence to point 27, and thusare ata voltage which is slightly less positive than emitter 22 of transistor -16 and emitter 23 of transistor 17.

This initial condition establishes a small forward bias on transistors 16 and 17. Their bases are. slightly negativewith respect to their emitters, which allows a small amount of current to flow fromemitter 22 to the col lec tor of transistor, 16, andfrom emitter 23 to .the; collector f. rans t 7- Olrrent furnished by battery 19 will enter the primary windings 12 and 13 of transformer 10 at point 24, and after passing through transistors 1 6 and 17 will return to the battery via point 18.

Normally there would be a small circuit unbalance which would result in a difference in the degree of forward conduction of transistors 16 and 17, thereby causing proper feedback to be developed, thus initiating oscillations.

First half cycle of operation In the following discussion it is assumed that transistor 16 begins to conduct harder than transistor 17. As it does so, the voltage at emitter 22 which had an initial value of +12 volts becomes slightly less positive due to the induced voltage drop in the primary winding 12 caused by the small increase in the forward current through transistor 16.

Due to transformer action and the polarity of the feedback winding with respect to the primary winding 12, there is also a small decrease in the voltage present at base 20 of transistor 16.

The base 20 remains negative with respect to emitter 22. Thus the small initial increase in forward conduction is aided with positive feedback furnished by the feedback winding 14. Transistor 16 conducts more and more heavily in a regenerative fashion.

After a very short time transistor 16 reverts to a saturated condition in which the forward resistance becomes extremely low. In this condition the base 20 is at a potential of .5 volt, and emitter 22 is at a potential of +.5 volt. Thus there is a forward bias of +1.0 volt.

At this time battery 19 is connected directly across the series combination of primary winding 12 and transistor 16. Nearly all of the supply voltage is dropped across primary winding 12.

Thus the upper end of primary winding 29 of step-up transformer 30, which is connected to emitter 22 of transistor 16, is at a potential of +5 volt.

While positive feedback is driving transistor 16 into saturation, negative feedback is driving transistor 17 into cutoff. This is accomplished in the following manner.

As emitter 22 of transistor 16 is being driven in the negative direction to +5 volt, the increasing current through primary winding 12 induces a voltage in primary winding 13, due to auto-transformer action, which causes the potential on emitter 23 of transistor 17 to become more positive.

The increasing current through primary winding 12 also induces a voltage in feedback winding 15 through transformer action. Thus the potential on base 21 of transistor 17 is increased.

When transistor 16 is in the saturated or fully conducting condition, transistor 17 is cut off for the following reasons.

The emitter 23 of transistor 17 attains an end potential of +24 volts, and the base 21 of transistor 17 attains an end potential of +30 volts. The base 21 potential is more positive than the emitter 23 potential due to the step-up ratio which exists between the feedback winding 15 and primary winding 13. Thus negative feedback is applied to transistor 17, and it remains cut off.

Since the lower end of primary winding 29 of step-up transformer 30 is connected to emitter 23 of transistor 17 it is at a potential of +24 volts. Thus there is an instantaneousvoltage across the primary winding 29 of transformer 30 which very nearly equals twice the battery 19 supply voltage of 12 volts.

The circuit remains in the stable condition described above for a length of time determined by certain circuit parameters. These include the size and type of core material used in transformer 10 and its magnetic characteristics, the number of turns in primary windings 12 and 13, and the battery 19 supply voltage.

Thefrequencyof operation of one model power, sup:

ply was in the order of 1000 cycles per second for a battery 19 supply voltage of 14 volts.

The first half cycle of circuit operation has been cornpleted, and after the predetermined length of time the rate of change of current flow through the primary winding 12 is reduced to zero. At this time the rate of change of magnetic flux in the core 11 of transformer is also reduced to zero, and the magnetic field about core 11 begins to collapse. This event signifies the beginning of the second half cycle of circuit operation.

Second half cycle of operation Examining transistor 16 and its associated circuit potentials upon collapse of the magnetic field, it is found that the voltage induced in primary winding 12 causes the potential on emitter 22 to increase in the positive direction.

The collapsing magnetic field also induces a voltage in feedback winding 14 which causes the potential on base 20 to increase in the positive direction.

The emitter 22 of transistor 16 attains an end potential of +24 volts, and the base 20 of transistor 16 attains an end potential of +30 volts. The base 20 potential is more positive than the emitter 22 potential due to the step-up ratio which exists between the feedback winding 14 and primary winding 12. Thus negative feedback is applied to transistor 16, and it reverts to a cutoff condition.

As this happens, the potential at the upper end of primary winding 29 of step-up transformer 30, which is connected to emitter 22 of transistor 16, is raised from +.5 volt to +24 volts.

While negative feedback is driving transistor 16 from conduction into cutoff, positive feedback is driving transistor 17 from cutoff into conduction. This is accomplished in the following manner.

The collapse of the magnetic field previously mentioned induces a voltage in primary winding 13 which causes the potential on emitter 23 of transistor 17 to become less positive. The field collapse also induces a voltage in feedback winding such that the potential on base 21 of transistor 17 becomes less positive.

The emitter 23 attains an end potential of +5 volt, and the base 21 attains an end potential of -.5 volt. Since at this time there exists a forward bias of +1.0 volt on transistor 17, it is driven into heavy conduction or saturation. The resultant increase in flow of current from battery 19 through primary winding 13 and transistor 17 back to point 18 causes an induced voltage drop across primary winding 13 which tends to aid in lowering the potential on emitter 23 of transistor 17. Through transformer action and the polarity of the feedback winding 15, the induced voltage in winding 15 tends to aid in lowering the potential on base 21 of transistor 17. Thus positive feedback furnished by the feedback winding 15 aids conduction by transistor 17.

Since the lower end of primary winding 29 of step-up transformer 30 is connected to emitter 23 of transistor 17, it also'attains an end potential of +5 volt. Thus there is an instantaneous voltage across the primary winding 29 of transformer 30 which very nearly equals twice the battery 19 supply voltage of 12 volts.

The circuit remains in the stable condition described above for a length of time determined by the previously mentioned circuit parameters. After this interval the rate of change of current flow through the primary winding 13 is reduced to zero, and the rate of change of magnetic flux about core 11 of transformer 10 is also reduced to zero. The magnetic field about core 11 begins to collapse, signifying the completion of the second half cycle of circuit operation.

The time required for the circuit to transfer from the first stable condition to the second stable condition is extremely fast. It is a function of circuit parameters including the characteristics of the feedback transformer 10, the switching speeds offered by transistors 16 and 17, and the battery 19 supply voltage. The rise and fall times of the square wave voltage which is present across the two emitters 22 and 23 of transistors 16 and 17, respectively, was in the order of 6 microseconds in one model power supply which was constructed.

This square wave voltage can be stepped up to any desired value by means of a suitable turns ratio between secondary Winding 31 and primary winding 29 of transformer 30. It could then be applied to a suitable semiconductor type rectifier network such as the full wave bridge configuration shown in Fig. 1, consisting of diodes 32, 33, 34, and 35. The resulting direct current output could be filtered and applied as plate supply voltage to most any electronic device.

While there has been shown and described what is at present considered to be the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made therein without departing from the scope of the invention as defined in the appended claims.

Having disclosed my invention, I claim:

1. A pulse generator comprising, in combination, a pair of switching transistors each having an emitter and a base and a collector, said transistors being arranged in the common collector configuration, a feedback transformer having a saturable toroidal core and a bifilarly wound primary connected between the emitters of the switching transistors and a bifilarly wound secondary connected between the bases of said transistors, a source of direct current connected between the midpoint of said primary and said collectors, and signal output means coupled to said emitters, each of said bifilarly wound primary and secondary comprising two portions wound in the same direction.

2. A transverter in accordance with claim 1, in which said signal output means comprises a non-saturated output transformer having a primary connected in shunt relation to said emitters and a secondary.

References Cited in the file of this patent UNITED STATES PATENTS 2,748,274 Pearlman May 29, 1956 2,843,815 Driver July 15, 1958 2,852,730 Magnuski Sept. 16, 1958 2,883,539 Bruck et al. Apr. 21, 1959

Images